A method and system for denoising artificial source electromagnetic prospecting signals with noisy reference traces

Abstract

The invention discloses a method and system for denoising an artificial source electromagnetic prospecting signal with a noise reference track. A noise observation point is set up in the observation area, and the noise observation point only collects a noisy magnetic field signal; the denoising method specifically includes: Based on orthogonal matching pursuit, the observed noise signal is sparsely decomposed; according to the decomposition result of the noise signal, the signal-to-noise separation is performed on the observed signal. The present invention only collects pure noise signals by adding a "pure" noise observation device, utilizes the periodic characteristics of artificial source signals, and based on the discrete characteristics of the frequency spectrum, transforms the denoising problem into a noise solving problem, reduces the difficulty of denoising, and improves the denoising success rate.

Description

technical field [0001] The invention belongs to the technical field of artificial source electromagnetic prospecting, in particular to a method and system for denoising artificial source electromagnetic prospecting signals with noisy reference traces. Background technique [0002] The statements in this section merely provide background information related to the present disclosure and do not necessarily constitute prior art. [0003] Due to the characteristics of controllable field source, known signal type and relatively high signal-to-noise ratio, the artificial source electromagnetic prospecting method has been widely used in the fields of energy, resource exploration, geological disasters and urban geological survey. Electromagnetic method (CSAMT), wide field electromagnetic method (WFEM), etc. [0004] However, with the continuous development of human society in recent years, strong human noise interference is ubiquitous and has become one of the key factors affecting...

AI Technical Summary

Problems solved by technology

The inventors found that among them, the statistical method adopts the method of superposition estimation to realize signal-to-noise separation, but it needs a large number of samples as a premise, and it is difficult to achieve effective denoising when the sample data is insufficient; the filtering method is mainly based on the noise and signal in the frequency spectrum There is a significant difference between the two in the middle or after the signal is projected into a specific space, so as to achieve signal-to-noise separation. If the spectral characteristics of the signal and noise are complex and overlap each other, it is difficult to separate them in the corresponding space. For example, in the face of some In the case of unsteady electromagnetic noise, effective denoising processing is difficult to complete; the empirical mode method can deal with signal noise in a variety of complex situations, and the signal-noise separation effect is remarkable, but when the effective signal and noise signal spectral components overlap seriously However, there are still great difficulties in achieving effective signal-to-noise separation, and such methods are mostly used in natural source electromagnetic exploration, while the properties of artificial source electromagnetic signals and magnetotelluric signals are very different, and the spectral discrete characteristics of artificial source effective signals often affect the noise effect

Images